INDEX
    Explanations

    comparisons

    New Auto-Interp
    Negative Logits
     bonté
    -0.68
    /**
    -0.68
    +#+#
    -0.68
     trône
    -0.65
     avoient
    -0.65
     trouvera
    -0.63
    AccessorTable
    -0.63
    AndEndTag
    -0.63
    ]--;
    -0.63
     fourrure
    -0.63
    POSITIVE LOGITS
     a
    0.56
     way
    0.50
    Hentet
    0.45
     ways
    0.45
     terms
    0.44
     style
    0.43
     pace
    0.43
     place
    0.42
     space
    0.42
     fashion
    0.40
    Act Density 0.003%

    No Known Activations